Dr. Grahame Holmes

For 26 years, Dr. Holmes was a faculty member at Monash University, working in the area of Power Electronics, where he established the Power Electronics Research Group to support graduate students and research engineers working together on a mixture of theoretical and practical R&D projects. In 2010, Professor Holmes moved with his group to take up the position of Innovation Professor – Smart Energy Systems at RMIT University, allowing him to broaden his research activities into a diversity of applications of power electronics, including particularly Smart Grids and Smart Energy technologies.

Professor Holmes has a strong commitment and interest in the control and operation of electrical power converters. He has made a significant contribution to the understanding of PWM theory through his publications and has developed close ties with the international research community in the area. He is a Fellow of the IEEE, has published over 200 papers at international conferences and in professional journals, and regularly reviews papers for all major IEEE transactions in his area. His research interests include Smart Grid Technologies and Applications, Power Electronic Converter Modulation and Control,  Resonant Converters, Multilevel Converters, and much more.

Dr. Holmes was highly involved in the beta testing of DSIM and therefore the first large-scale adopter of the revolutionary simulation engine.

“A major challenge for investigating grid-connected inverters is to simulate the complete inverter operation over extended time periods. Conventionally, this is done either by using averaged simulation models or by a series of short-term “snapshot” simulations of specific transient events. Neither approach is satisfactory, and simulation times are often substantial. The benefit of DSIM to address this issue was its very fast simulation execution time compared to other simulation packages.

To begin, a grid inverter was modeled in extreme detail, incorporating all physical components of the experimental system and its measurement circuits. Modeling at this level of detail avoids mistakes and errors that are often created using reduced or averaged circuit models. For example, filter time constants of the physical system are automatically correct since they derive directly from the represented components instead of from a time constant calculation. Next, almost all experimental converter controller software was implemented into a C block. It is important to minimize any editing away from the experimental system software to avoid approximation, interpretation, and possible errors. Eventually, with considerable effort, over 95% of the complete experimental system software was included in the simulation and executed without error.

Taken together, the resulting simulation is an almost exact representation of the actual physical hardware and software and executes at sufficient accuracy to allow much of the software algorithms to be commissioned using simulation before proceeding to experimental validation. Furthermore, even commissioning issues that must be resolved in hardware can be explored more quickly and flexibly than is the case using more conventional approaches.

DSIM executes this system in about 1/15th real-time. Or, in other words, 10 seconds of simulation time executes in about 150 seconds. This is sufficiently fast to allow useful investigations to proceed and thus makes simulating at this level of detail on a laptop computer viable.

The approach has been used both for grid-connected inverter investigations, and to simulate the operation of a laboratory-level dc-synch machine set and its controlling converters. Currently, work is underway to integrate these simulations into a mixed micro-grid system, and then to investigate machine/inverter oscillatory interactions over several 10’s seconds. I know of no other simulation package that can manage this level of simulation using a laptop computer.“